This study considers the real-time construction and visualisation of the hydrodynamic field in physical experiments of liquid sloshing. An intelligent image processing method (IPM) integrated with the artificial damping model (ADM) is proposed to achieve this purpose. Through image processing, kinematic behaviours of the sloshing liquid are extracted. Deep neural networks (DNN) are established to recognize the wave-breaking thresholds, followed by the process of direct linear transformation (DLT) and contour detection to extract and reconstruct the free surface with bases of polynomials. A mathematical linkage between the targeted sloshing dynamics and observed kinematic behaviours is derived based on the potential-flow theory, where the artificial damping model is specially introduced to equivalently reflect the important viscous dissipation effect. Through the above IPM-ADM, real-time pressure fields in the sloshing tank can be visualised directly from experimental videos. A series of physical experiments are newly conducted to further validate the effectiveness of the IPM-ADM, with particular attention on resonance and near-resonance cases. Good agreements can be found between corresponding pressure histories of the IPM-ADM predictions and physical measurements. Characteristics of the pressure fields are also visualised and discussed.